Computing devices are often connected by network connections. For example, computing devices are frequently connected by Ethernet network connections. Such computing devices support a variety of capabilities (e.g., connection data rates, communication protocols, error correction protocols, etc. For example, some computing devices support forward error correction (FEC). When utilizing FEC, a sender of data encodes the message in a redundant manner using error correcting codes (ECC) to increase the chances that a receiver can decode the data without requesting a retransmission.
Some computing devices include a network device motherboard (e.g., a built-in network device motherboard or an adapter network device motherboard) that interfaces with a separate transceiver. For example, the network device motherboard may include and/or be communicatively coupled with a port that can receive a transceiver (e.g., different transceivers may provide different PHY types). The interface to the transceiver may be, for example, enhanced small form-factor pluggable (SFP+), quad small form-factor pluggable (QSFP), or any other type of interface for communicatively coupling a transceiver with a device (e.g., a network device in a computing device such as a server, a switch, etc.). Alternatively, a transceiver may be integrated with the network device motherboard that is integrated into a computing device and/or on an adapter installed in the computing device.
In some networking environments, network cabling between two networked devices may not provide sufficient communication fidelity. For example, communications transmitted over long distances may not be received properly when reaching their destination. In such instances, a re-timer (e.g., an Ethernet re-timer) may be connected intermediate between the two networked devices. The Ethernet re-timer includes the capability to perform Ethernet link training, recover a clock, and re-drive data to extend electrical drive capabilities beyond what a typical single driver can accomplish. The re-timer resides at an intermediate point between the two network devices, effectively breaking one long distance communication into two electrically shorter connections.